Few-Shot Learning for Rooftop Detection in Satellite Imagery

Deep Learning Tutorial

Giorgio Coppala, Nadine Daum, Elena Dreyer, Nico Reichardt

Policy Relevance

  • Many policy domains face limited labeled data
    (annotation is expensive, slow, or requires domain expertise)

  • Applications:

    • medical sector: rare disease detection
    • emergency management: flood extent mapping
    • climate & energy: solar PV rooftop assessment
    • urban planning: building footprints & infrastructure mapping

  • Few-shot learning (FSL) can help:

    • Learns to generalize from 1–5 labeled support examples per class
    • Learns a feature embedding and constructs class prototypes
    • Enables segmentation in a new city with minimal additional annotation

Problem Setting

  • Tutorial Task: apply Prototypical Networks to rooftop semantic segmentation using only a few labeled tiles

  • Few-shot segmentation helps the model capture key rooftop patterns from just a few Geneva tiles and apply them across the city

  • Demonstrates how rooftop maps can be generated to support solar potential analysis in a new city

Geneva outline with sample rooftop data (self-made visualization)

Dataset: Geneva

  • Size: 1,050 labeled image-mask pairs

  • Task: Binary segmentation masks (rooftop vs background)

  • Geographic splits: 3 grids/ neighborhoods (North, Center, South)

  • Image size: 250x250 pixels

  • Categories: Industrial, Residential

Inside the dataset

Geneva Animation: raw image → rooftop mask → overlay